The SAGE Handbook of Persuasion

by James Price Dillard

  High PC

  Professor Griffin, can I talk with you about the term paper assignment? I really love the topic you suggested for my paper; it’s so interesting and I’m learning so much! I’ve found a tremendous amount of material on this topic. I want to master all the material and write a really excellent paper, but to do that, I’m going to need a little more time than you originally allowed. You told us that we should make this paper something that we are proud of, and I really want to do that. So do you suppose I can turn the paper in next week? I know the extra time will allow me to produce an excellent paper, and one I think you’ll enjoy reading.

  Comforting (Relationship Breakup)

  Low PC

  Ben broke up with you? He’s an idiot! But, this isn’t the end of the world, you know. I mean, it’s not the worst thing that could happen to you, and to be honest, I think you’ll be better off without Ben. Anyway, there are tons of cute guys on this campus, you know, lots of fish in the sea. You just gotta get out there and catch another one! Keep in mind that no guy is worth getting all worked up about. I mean, it’s just not that big a deal, not at this point in life. You can do a lot better than Ben. Just remember that Ben isn’t worth any heartache and you’ll stop being so depressed about the whole thing.

  High PC

  Barb broke up with you? Oh man! I’m really sorry; I know you must be hurting right now. Do you want to talk about it? You were together a long time and were really involved with her, so you must have some real heartache. This just sucks; I’m really sorry, man. The same thing happened to me last year, and I remember how rotten it makes you feel. It’s especially tough when it’s sudden like that. It’s probably gonna take some time to work through it –after all, breaking up is a really hard thing. I know it may not mean very much right now, but keep in mind that you’ve got some good friends here –people who really care about you. I’m here whenever you want to talk about things.

  Notes: I obtained these messages from my advisor, Brant Burleson, early in my doctoral program at Purdue. I will forever be grateful for his influence on my thinking as a scholar of human communication.

  The second class of effects can be characterized as message outcomes (MO), which includes a range of effects happening after the message. Persuasive messages are generally created to change or stabilize attitudes (or determinants of attitudes such as beliefs and evaluations, see Fishbein & Ajzen, 1975), behavioral intentions, and actions, whereas supportive messages generally aim to generate cognitive (e.g., appraisals), affective (e.g., emotions), and/or behavioral (e.g., coping) change (see Burleson, 2009). Backing research that finds ME is a function of message characteristics, advice is more likely to be implemented and, thus, lead to desired change (e.g., assist in coping) if it is efficacious, feasible, and absent of limitations (MacGeorge et al., 2008). Likewise, high person-centered messages are more persuasive (e.g., Waldron & Applegate, 1998) and are shown to enhance emotional improvement (Jones & Guerrero, 2001) compared to low person-centered messages; in the context of emotional support, low person-centered messages can further exacerbate stress reactivity (Bodie, in press-b).

  A primary strategy in both persuasion and supportive communication research has been to focus on ME, a strategy that makes two primary assumptions: (1) whatever affects ME will similarly affect MO, and (2) ME is a causal antecedent of outcomes. Synthesizing relevant persuasion research to date, Dillard, Weber, and Vail (2007) found that the association between ME and MO was substantial (r = .41). Dillard, Shen, and Vail (2007) subsequently reported five studies that consistently supported ME as a causal antecedent of MO. A recent two-study report found similar results for the relationship between supportive message evaluations and outcomes (Bodie, Burleson, et al., 2012). Overall, then, it appears that a strong connection between evaluations and outcomes exists in at least two communication contexts, and it seems reasonable to assume that this association extends to other communication contexts as well.

  This finding has important methodological and conceptual implications. Although assessing ME is generally much easier, safer, and cheaper than is assessing MO, outcomes are generally of greatest practical interest. Practitioners of persuasion and support want to know what to say in order to change the attitudes and feeling states of others, not just whether message features are perceived to be better. Of course message evaluations are interesting and important in their own right, and there exist a host of potential evaluations that can be made of supportive and persuasive messages. Likewise, if we were to focus on outcomes to the exclusion of effects such as perceived comprehension known to influence these outcomes, we would miss out on important opportunities to advance theory and practice (see Berger, 1989).

  Beyond Effects: How Messages “Work” Within Interaction

  * * *

  However useful theories of supportive messages have been in explicating the features that are evaluated more positively and typically help or hinder outcomes, only recently have these theories been incorporated into more comprehensive theories of how messages “work” (i.e., have their effects). One of the earliest attempts was Burleson and Goldsmith’s (1998) theory of conversationally induced reappraisals, which proposes that successful comforting messages work by changing the implications of the person-environment relationship for the recipient’s coping ability and general well-being (Jones & Wirtz, 2006). In a similar manner, influence scholars have shown the utility of appraisal-based theories for the study of how influence messages operate to create certain emotional responses (Dillard, Kinney, & Cruz, 1996). In contrast to strategy-based frameworks that attempt to match messages to “a predefined and static environmental circumstance,” appraisal theories suggest that messages work by “discursively constructing useful appraisals of particular person-environment configurations” (Burleson & Goldsmith, 1998, p. 259). Thus, an important contribution of appraisal theories is to shift our focus from static message effects to the interactions in which messages are more or less efficacious.

  Supportive and Persuasive Interactions

  Supportive and persuasive interactions are most commonly described using phase models. For instance, the interactive coping model forwarded by Barbee and her colleagues (Barbee, 1990; Barbee & Cunningham, 1995; Barbee et al., 1993; Derlega, Barbee, & Winstead, 1994) outlines four phases of supportive interaction—support activation, support provision, target reactions, and helper responses—and persuasive interaction is discussed with somewhat similar phases (e.g., compliance solicitation, resistance; Sanders & Fitch, 2001). To date, however, only a handful of studies have actually investigated supportive or persuasive interactions in the ways implied by these models. Research suggests, for instance, that advice has more positive outcomes when directly solicited (e.g., Goldsmith, 2000) and when prefaced by highly person-centered emotional support and adequate problem analysis (Feng, 2009). Likewise, a number of studies have investigated “sequential request strategies” (D. J. O’Keefe, 2002) and general principles explaining how and why people change and adapt (or fail to do so) when initial persuasive attempts are unsuccessful (e.g., Hample & Dallinger, 1998). Nevertheless, much persuasive and supportive communication research utilizes methodological tools such as paper and pencil surveys that are far from “interactive” at their core (Jones & Guerrero, 2001; Wilson, 2002). The primary strategy used to explore support and persuasion as interactive phenomena is to explore how people produce and process messages in the service of interaction goals.

  Message Production

  Dillard and Solomon (2000) describe four message production processes, namely situation comprehension, goal formation, planning, and behavioral enactment. To these processes, Burleson and Planalp (2000) suggest two others, monitoring and reencoding. Thus, variability in the content and quality of influence or supportive messages can be explained as a function of abilities to define the situation, form relevant intentions to achieve some purpose, generate cognitive representations of action plans, carry out the action plan(s), observe a
nd evaluate the outcomes of the generated action, and, if necessary, alter goals, plans, and actions. Given that most research to date has focused on the first two of these processes, the sections that follow are necessarily limited, and others are encouraged to explore the contributions of support and persuasion to planning, enactment, monitoring, and encoding processes.

  Situation Comprehension

  In general, people who view their social milieu in more complex ways are better able to produce messages that take into account a greater number of contextual features and contingencies. Interpersonal cognitive complexity (ICC) is a stable individual difference that represents this ability to comprehend and process social information, including messages (Burleson & Caplan, 1998). Perceivers with higher ICC have more differentiated, abstract, and organized constructs (or schemas) for processing social information and, thus, have more advanced social perception skills than do less complex perceivers. Considerable research indicates that ICC is positively associated with the ability to generate sophisticated supportive and persuasive messages (for review, see Burleson & Caplan, 1998). Thus, ICC is a domain specific general ability to “size up” people and situations (Burleson & Bodie, 2008). Although domain is usually not cast at the level of communicative context, it is possible that interpersonal influence and supportive contexts constitute domains of knowledge. Indeed, it is easy to imagine a person who has a high level of ICC (i.e., can “size up people”) but little knowledge of how attitude or affect change occurs.

  Other variables that seem to describe ways in which individuals are likely to “size up” persuasive and supportive interactions can be classed into what scholars call implicit theories or how individuals structure their beliefs about a particular phenomenon. Burleson and Planalp (2000), for instance, discuss implicit theories of emotion—how the individual thinks emotion works—which seem highly relevant to the supportive context. Indeed, the support attempts of individuals who believe emotions cannot be changed or that they dissipate over time irrespective of communicative efforts should vary drastically from those of individuals who see emotions as pliable and able to be manipulated through talk. Similarly, B. J. O’Keefe (1988) discusses implicit theories of communication, or how the individual structures the workings of communication; similar ideas have been forwarded about persuasion (Roskos-Ewoldsen, 1997) and other elements of communication (e.g., listening; Bodie, St. Cyr, Pence, Rold, & Honeycutt, 2012). In general, implicit theories offer an explanatory link between goals and message outcomes; in other words, two individuals may have a similar goal in mind (e.g., to gain compliance from a friend) but produce drastically different messages in order to obtain that goal because each views the purpose of communicating differently in that situation (see also Planalp & Knie, 2002).

  Goal Generation

  The specific nature of goals—“future states of affairs that an individual is committed to maintaining or bringing about” (Dillard et al., 2001, p. 433)—has been a standard focus of both persuasion and supportive communication research. Interpersonal influence scholars typically talk about primary and secondary goals (Dillard, 1990) with primary goals (e.g., gaining assistance) referring to goals that motivate interpersonal influence, and secondary goals (e.g., identity goals) those that constrain behavior. In general, to the extent that putative support providers have “a desire to modify the target’s behavior,” they hold a primary goal similar to that of compliance seekers. Interestingly, although “give advice” is often listed as a primary influence goal (for review, see Wilson, 2002, Table 5.1), to “be supportive” is usually listed as a secondary goal (Meyer, 2004).

  For the support context, two primary goals have been identified. Individuals are said to primarily be emotion- or problem-focused, though a variety of subgoals under each category have been documented (e.g., MacGeorge, 2001) as have various ways in which these goals might be accomplished (e.g., avoid embarrassing talk about emotions versus given her the right advice about how to solve the problem; Burleson & Gilstrap, 2002). In addition, scholars recognize several secondary goals of support, including resource goals (e.g., the relationship may suffer if support seeking and provision are imbalanced in a relationship), arousal goals (e.g., the need to protect one’s own emotional well-being), and interaction goals (e.g., the need to protect the other’s face; Goldsmith, 1992). Indeed, the need to accomplish multiple, often competing, goals is a fundamental characteristic of supportive (Goldsmith, 1992), compliance gaining (Wilson, 2002), and interpersonal influence (Dillard, 1990) interactions.

  Regardless of interaction context, a focus on goals provides scholars with a “highly proximal generative mechanisms for [explaining] the messages people produce” (Burleson & Gilstrap, 2002, p. 44). Indeed, the “sine qua non of goal categories … is their ability to conjoin otherwise isolated actions into coherent, connected, rule-guided (and in some cases sequential) patterns” (Hoffman, Mischel, & Mazze, 1981, p. 212). Moreover, invoking the notion of goals allows scholars to explain how variables such as ICC impact message production ability—individuals who are more cognitively complex see communicative situations in more complex ways and are thus better able to generate more complex goals for social interaction and think about how to integrate seemingly inconsistent goals (e.g., advise and save face; see Dillard, 1997, for an extended discussion of the benefits of the goal construct).

  What still remains somewhat elusive, however, is exactly how goals arise and change over the course of interaction. Wilson’s (1990) cognitive rules (CR) model is a general theory of goal formation that has been applied in the context of interpersonal influence with similar implications for supportive interactions. The CR model proposes that goal-relevant knowledge is stored in long-term memory within nodes that represent concepts (e.g., people, traits, settings). Cognitive rules link various interaction goals (e.g., give advice) with these various concept nodes. So, when faced with a situation that calls forth a primary goal and other situational features that make salient various secondary goals, the CR model proposes that certain cognitive rules are activated and, thus, are more likely to be employed than other, less salient cognitive rules. This might help explain why, for instance, men tend to employ less helpful forms of support, especially with other men—although a primary goal of helping his friend work through his emotions “pushes,” a secondary identity or relational resource goal might “pull” back certain (more effective) strategies (cf. Burleson, Holmstrom, & Gilstrap, 2005).

  Finally, although this review has focused solely on message production abilities, research clearly shows that motivation is an equally important contributor to effective message production (Berger, 1996). Burleson, Holmstrom, and Gilstrap (2005) identified three types or components of motivation for supportive message production: “goal motivation, or the desire to achieve a particular social outcome (such as relieving the emotional distress of another); effectance motivation, or the level of confidence a person has in his or her ability to achieve an outcome (such as improving another’s affective state); and normative motivation, or the desire to behave in role-appropriate ways (such as saying the correct things when comforting others, according to salient social norms)” (pp. 470–471). Interestingly, a focus on ability and motivation is also evident in the literature on message processing, and it is to that literature we now turn.

  Message Processing

  Researchers studying supportive communication have borrowed extensively from the persuasion literature to explain variability in message effects. Scholars interested in variability in advice receptiveness have found that advice is better received when offered by more expert and relationally close individuals who have a stronger influence history with the recipient (Feng & MacGeorge, 2006); these variables map conceptually to credibility, similarity, and authority found to improve the likelihood of persuasive success (see D. J. O’Keefe, 2002). Likewise, although substantial research indicates that highly person-centered comforting messages are evaluated more positively and have more positive outc
omes than less person-centered messages, growing evidence indicates that the effect of PC is moderated by several demographic factors (e.g., recipient’s ethnicity), personality traits and cognitive factors (e.g., ICC), and features of the interactional context (e.g., recipient’s need for support; for review, see Bodie & Burleson, 2008). Recently, my research has focused on exploring the degree to which these moderating variables can be more succinctly organized under a “dual-process” framework. The primary contribution of this work is to suggest message effects are a joint function of message content and how these messages are attended to and processed.

  The General Dual-Process Framework

  Dual-process theories, regardless of whether they are directed at explaining person perception, stereotyping, decision making, or communicative functions, assert that information processing lies on a continuum, with “[the] anchors of this continuum [reflecting] the ‘duality’ invoked by these” theories (Moskowitz, Skurnik, & Galinsky, 1999, p. 13). These anchors have been labeled mindless/mindful, top-down/bottom-up, automatic/intentional, effortless/effortful, holistic/analytic, and systematic/heuristic, just to name a few. But these dichotomies often obscure the notion that processing lies on a continuum, and, thus, varies in extent as opposed to being simply high or low. In my work, I have chosen to refer to this variation as elaboration in line with a popular dual-process theory of persuasion (Wegener & Claypool, 2000).

  In general, as elaboration increases, recipients begin to more carefully reflect on the content of issue-relevant information, thoughtfully consider this information in relation to prior ideas, and give closer attention to multiple aspects of the communicative context. When elaboration is high, change is said to occur through “high thought” mechanisms. As elaboration decreases, cues not directly related to the issue at hand begin to have a larger influence on communication outcomes through mechanisms of change that require relatively little thought. With regard to explaining variation in message outcomes in particular, dual-process theories postulate that (1) multiple factors influence the amount of elaboration accorded by recipients, and (2) the effects of communicative structures (i.e., source, message, context, recipient) on outcomes vary as a function of elaboration (see Holmstrom et al., 2011). In particular, dual-process theories hold that message content has the strongest effect on outcomes when recipients are both motivated to attend to the message and possess the ability to consider its content thoughtfully.